From: David Miller <da...@davemloft.net> Date: Thu, 17 May 2018 14:41:17 -0400 (EDT)
> From: Jose Abreu <jose.ab...@synopsys.com> > Date: Thu, 17 May 2018 14:24:42 +0100 > >> Given that the difference between better/worst is < 1%, I think >> we can conclude patches 3-13 don't affect the overall >> performance. I didn't profile the cache hits/miss though ... > > Ok, thanks for making an effort to look into this more thoroughly. > > I'll apply this series to net-next, thank you. Sorry, I had to revert. It is one thing to say that you lack the hardware to physically test the changes on all chip types. It is yet another to not even test the build properly: drivers/net/ethernet/stmicro/stmmac/dwmac-sun8i.c:494:10: error: initialization from incompatible pointer type [-Werror=incompatible-pointer-types] .init = sun8i_dwmac_dma_init, ^~~~~~~~~~~~~~~~~~~~ drivers/net/ethernet/stmicro/stmmac/dwmac-sun8i.c:494:10: note: (near initialization for ‘sun8i_dwmac_dma_ops.init’) cc1: some warnings being treated as errors